Evaluation method for echelon utilization power battery

Abstract

The invention relates to the technical field of battery pack evaluation, and discloses an echelon utilization power battery evaluation method, which comprises the following steps: performing appearance and smell inspection, and screening batteries without damage, liquid leakage and peculiar smell; safety performance tests are sequentially carried out, and the safety performance tests comprise insulation tests, voltage-withstanding tests and elimination of unqualified batteries; performing dynamic environment simulation, performing health state detection through room-temperature and high-low-temperature charging and discharging, rate testing and internal resistance analysis, and evaluating residual capacity and consistency; a high-precision universal meter is adopted for voltage consistency evaluation, the resolution ratio is 0.1 mV, and the measurement error is smaller than or equal to + / -0.5 mV; the temperature consistency is detected by a distributed temperature sensor, the precision is + / -0.5 DEG C, and the sampling interval is less than or equal to 2s; bMS function verification is carried out, and SOC estimation and voltage / current measurement precision are calibrated; and integrating the detection data, and generating a battery health state report and echelon utilization suggestions. The detection efficiency and safety are improved, the resource utilization rate is maximized, and the cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of battery pack evaluation, and particularly to a method for evaluating second-life power batteries Background Art

[0002] In recent years, with the global high emphasis on environmental protection and sustainable development, the new energy vehicle industry has shown a rapid development trend. As the core component of new energy vehicles, the market demand for power batteries has also increased sharply. However, with the popularization of a large number of new energy vehicles, power batteries gradually retire after a certain usage cycle. How to properly handle these retired power batteries has become an important issue faced by the industry. Second-life utilization, that is, the reasonable evaluation and reuse of retired power batteries, is regarded as an effective way to solve this problem and has become a research hotspot in the industry.

[0003] In the existing technology, there are many problems in the evaluation of retired power batteries.

[0004] Firstly, the detection items are not comprehensive. Most of the existing evaluation methods focus on the attenuation of battery capacity, but seriously ignore the detection of safety performance, such as key safety indicators of insulation performance and withstand voltage performance. At the same time, the accuracy detection of the battery management system (BMS) is not emphasized either, which poses a great safety hazard during the secondary utilization process.

[0005] Secondly, the standardization degree is insufficient. The current detection process lacks a unified standard specification, and there are significant differences in the test conditions and equipment accuracy adopted by different detection institutions or enterprises, resulting in the lack of comparability of detection results and making it difficult to objectively and accurately evaluate retired batteries.

[0006] Moreover, the detection efficiency is low. Traditional detection methods mainly rely on manual operation, with extremely low automation. Especially in complex test environments, such as high and low temperature environments and large-rate charge and discharge, the test takes too long, seriously affecting the processing efficiency of retired batteries and the process of second-life utilization.

[0007] In addition, the data management is also relatively weak. The detection data records are discontinuous and incomplete, and a comprehensive and effective battery state of health (SOH) evaluation system cannot be formed, which is not conducive to the scientific and reasonable screening and reuse of retired batteries.

[0008] Although there are already relevant studies on the cascade utilization of power batteries, there are still many prominent problems. For example, a method for sorting and evaluating the cascade utilization of power batteries disclosed in Patent CN201210267131.7 only focuses on the appearance, performance parameters, and internal structure of power batteries. Its operation process is not unified and standardized, resulting in differences in the evaluation results of different operators and inconsistencies with the actual usage effects. Another method for screening the cascade utilization of retired batteries disclosed in Patent CN202510068090.6, although involving automated screening, can only complete static screening and cannot comprehensively and accurately restore the working state of power batteries, leading to deviations between the screening results and the actual situation.

[0009] In summary, the existing technologies do not systematically integrate multi-dimensional detection indicators and lack customized evaluation criteria for cascade utilization scenarios, resulting in insufficient reliability in the screening and reuse of retired batteries. There is an urgent need for a more perfect, efficient, and accurate method for evaluating power batteries for cascade utilization to meet the development needs of the industry. Summary of the Invention

[0010] The present invention aims to provide a method for evaluating power batteries for cascade utilization to solve the problem of insufficient reliability in the screening and reuse of retired batteries due to the lack of customized evaluation criteria for cascade utilization scenarios.

[0011] To solve the above problems, the present invention adopts the following technical solutions: A method for evaluating power batteries for cascade utilization includes the following steps: Step 1: Conduct appearance and odor inspections to screen out batteries without damage, leakage, or abnormal odor. Step 2: Conduct safety performance tests in sequence. The safety performance tests include insulation, withstand voltage tests, and ultrasonic scanning for internal structure analysis to eliminate unqualified batteries. Step 3: Dynamic environment simulation. Through room temperature, high and low temperature charge and discharge, rate tests, and internal resistance analysis, health state detection is carried out. At the same time, a multi-scenario working condition simulation test bench is connected to simulate constant power charge and discharge in the energy storage scenario with a fluctuation within ±10%, and pulse charge and discharge in the low-speed power scenario with a pulse width of 0.1 - 2 s and a peak current of 3C. Record the voltage transient response time and fluctuation amplitude to evaluate the remaining capacity and consistency. The voltage consistency is evaluated using a high-precision multimeter with a resolution of 0.1 mV and a measurement error of ≤±0.5 mV. The temperature consistency is detected by a distributed temperature sensor with an accuracy of ±0.5°C and a sampling interval of ≤2 s. In the high and low temperature charge and discharge tests, the high temperature environment is 50°C ± 1°C, and the low temperature environment is -20°C ± 1°C. Five charge and discharge cycles are carried out at each temperature point, and the average capacity attenuation rate is taken as the evaluation basis. Among them, when the low temperature discharge capacity attenuation rate ≤ 35%, it is determined to be qualified. Step 4: Conduct BMS function verification, calibrate the SOC estimation and the measurement accuracy of voltage and current; synchronously perform historical data traceability: analyze the charge and discharge logs of the battery in the 100,000 kilometers before retirement, extract the deep discharge frequency with SOC < 10% and the proportion of the high-temperature operation duration with temperature > 45°C, and calculate the battery stress accumulation index; Step 5: Integrate the detection data to generate a battery health status report; combine the stress accumulation index with the working condition simulation data to generate a scenario adaptation risk score within the range of 0 - 100 points; In the dynamic environment simulation, the temperature control accuracy of the high and low temperature chamber is ±0.5°C, and the temperature change rate ≥ 5°C / min, which can simulate extreme temperature conditions within the range of -30°C to 60°C.

[0012] The present invention accurately simulates the extreme environment of -30°C to 60°C (accuracy ±0.5°C, temperature change rate ≥ 5°C / min) through a high and low temperature chamber, combines the high-precision voltage (0.1mV resolution) and temperature (±0.5°C accuracy) consistency detection, can truly reproduce the actual working conditions of the battery, accurately identify the differences between single cells, timely capture abnormal temperature rise, improve the accuracy and scenario adaptability of the evaluation results, and ensure the safety and service life of the battery pack for secondary use. Conduct 5 charge and discharge cycles in the environment of 50°C ± 1°C and -20°C ± 1°C, and take the low-temperature discharge capacity attenuation rate ≤ 35% as the standard to quantitatively evaluate the low-temperature performance and cycle stability, accurately screen the batteries suitable for cold regions, eliminate accidental errors, and balance the usability and resource utilization rate. The present invention realizes the efficient and accurate evaluation of the safety performance, health status and BMS function of retired power batteries through multi-dimensional detection, automated processes, dynamic environment simulation and scenario-based grading standards, improves the detection efficiency and safety, maximizes the resource utilization rate and reduces the cost.

[0013] Preferably, the ultrasonic scanning internal structure analysis uses a multi-band focused ultrasonic probe with a central frequency of 5 - 20MHz, and the batteries with micro short circuits of the electrode sheet, lithium deposition thickness ≥ 50μm or electrolyte dry area > 5% are unqualified batteries.

[0014] Traditional methods cannot detect internal microscopic defects of the battery (such as lithium deposition and micro short circuits), resulting in the risk of thermal runaway after secondary use. The present invention uses a multi-band focused ultrasonic probe (5MHz for detecting the electrolyte distribution and 20MHz for scanning the electrode interface), and through acoustic impedance difference imaging, it can detect internal microscopic defects of the battery and avoid the risk of thermal runaway after secondary use.

[0015] Preferably, in the BMS function verification, the SOC estimation error is calibrated by simulating charge and discharge conditions. For the energy storage scenario, it should be ≤2%, and for the low-speed power scenario, it should be ≤5%. In the energy storage scenario, the voltage measurement error is ≤±5 mV, in the low-speed power scenario, the voltage measurement error is ≤±10 mV, in the energy storage scenario, the current measurement error is ≤±0.5% FS, and in the low-speed power scenario, the current measurement error is ≤±1% FS.

[0016] Calibrate the BMS accuracy by grading for different scenarios (energy storage SOC error ≤2%, voltage ±5 mV, low-speed power SOC error ≤5%, voltage ±10 mV), optimize system compatibility, avoid overcharge / overdischarge risks, improve the collaborative reliability between the battery pack and the application system, and promote BMS standard compatibility.

[0017] Preferably, in step three, the room temperature charge and discharge test is carried out in an environment of 25°C ± 1°C, the charge and discharge rate is 0.5C, the remaining capacity is calculated by the ampere-hour integration method, the error is ≤1.5%, and in the evaluation of voltage / temperature consistency, the cell voltage range is ≤50 mV, and the temperature range is ≤2°C.

[0018] Test the remaining capacity at a rate of 0.5C in an environment of 25°C ± 1°C (error ≤1.5%), control the voltage and temperature ranges (≤50 mV and ≤2°C), ensure accurate basic performance evaluation, reduce the risk of battery pack differences, extend the cycle life by more than 15%, and balance test efficiency and data stability.

[0019] Preferably, in step five, the calculation model for the scenario adaptation risk score is: R = 0.4×SOH + 0.3×(1 - |ΔV| / V_nom) + 0.2×(1 - K / 3) + 0.1×BMS_Accuracy Where SOH is the remaining capacity ratio, |ΔV| is the amplitude of the voltage fluctuation during the working condition simulation, K is the stress accumulation index, and BMS_Accuracy is the SOC estimation accuracy; When R ≥ 80 points, it is recommended for Class A energy storage scenarios, and when 60 ≤ R < 80 points, it is recommended for low-speed power scenarios. Through the calculation of the scenario adaptation risk score, appropriate scenarios can be accurately classified, making the cascade utilization of power batteries more reliable and targeted.

[0020] Preferably, in step two, in the insulation test, for the single cells in the energy storage scenario, a DC voltage of 1500 V is applied, the test duration is 90 s, and the insulation resistance should be ≥100 MΩ; for the battery modules in the low-speed power scenario, a DC voltage of 1000 V is applied, the test duration is 60 s, and the insulation resistance should be ≥50 MΩ.

[0021] The insulation test standard for energy storage single cells is 1500V, 90s, ≥100MΩ, and for low-speed power modules is 1000V, 60s, ≥50MΩ, taking into account safety and detection efficiency, eliminating instantaneous insulation fluctuations, and promoting the standardization of industry detection standards.

[0022] Preferably, in step three, the internal resistance analysis uses the AC impedance method, the test frequency range is 5mHz - 50kHz, the internal resistance is fitted through an equivalent circuit model, the measurement accuracy is ≤ ±0.8% or ±0.05mΩ, taking the larger value, and the internal resistance growth rate is calculated based on the initial internal resistance, with the threshold being ≤ 20% for the energy storage scenario and ≤ 30% for the low-speed power scenario.

[0023] Analyze the internal resistance in the frequency range of 5mHz - 50kHz (accuracy ±0.8% or ±0.05mΩ), quantify the aging degree with a growth rate ≤ 20% / 30%, distinguish the microscopic attenuation reasons, achieve accurate matching of aging warning and scenarios, and extend the cascade utilization cycle by more than 20%.

[0024] Preferably, the ultrasonic scanning internal structure analysis includes: for energy storage scenario batteries, the scanning resolution is ≥ 0.1mm², and it is determined as unqualified when the detected lithium deposition thickness is ≥ 80μm or the micro-short circuit point density is > 3 per cm²; for low-speed power scenario batteries, it is determined as unqualified when the detected dry electrolyte area is > 8% or the wrinkled area of the electrode sheet is > 5%.

[0025] Through ultrasonic scanning internal structure analysis, the internal analysis and judgment of power batteries can be carried out more accurately, internal defect batteries can be eliminated in advance, and the thermal runaway accident rate after cascade utilization is reduced by 40%.

[0026] Preferably, in the multi-scenario working condition simulation test bench: when simulating constant power fluctuations in the energy storage scenario, the voltage transient response time is ≤ 100ms and the fluctuation amplitude is ≤ 2% of the rated voltage; when simulating pulse charge and discharge in the low-speed power scenario, the SOC estimation offset is ≤ 5%, and the BMS temperature sampling delay is ≤ 500ms.

[0027] Accurately predict the failure mode of the battery in the actual scenario, and reduce the screening error by 35%.

[0028] Preferably, in the historical data traceability of step four: for batteries with a deep discharge frequency > 50 times and a high-temperature operation time reaching 20% of the total life, the stress accumulation index K ≥ 1.8; when K ≥ 2.0, generate a warning label of "Prohibited for high-rate scenarios".

[0029] Can reasonably utilize historical data and effectively reduce the prediction error of the cycle life.

[0030] The advantages of the present invention are: Significantly improved safety: Through insulation resistance testing, withstand voltage testing, and protection verification safety detection, retired batteries with potential safety hazards are accurately eliminated, reducing the risk of secondary utilization by more than 60%, and ensuring the application safety in scenarios such as energy storage and transportation.

[0031] Double optimization of cost and efficiency: The automated detection process reduces manual intervention by more than 70%, and the detection cycle is shortened from 48 hours of traditional methods to within 24 hours. At the same time, standardized equipment is used to reduce test errors, and the comprehensive cost is reduced by more than 30%.

[0032] Maximized resource utilization: A hierarchical mechanism of "direct utilization if up to standard - marginal re-evaluation - directional downgraded use - disassembly and recycling" is established. Batteries that do not fully meet the standards (such as remaining capacity of 40%-60%) are recommended to scenarios such as backup power supplies and small electronic devices, increasing the cascade utilization rate to 85% and reducing disassembly waste by 20% compared to traditional screening modes. Brief Description of the Drawings

[0033] Figure 1 It is a flowchart of the method of the present invention. Detailed Description of the Invention

[0034] The following is a further detailed description through specific embodiments: A method for evaluating the cascade utilization of power batteries, through an integrated multi-dimensional detection system such as safety performance detection, state of health assessment, and battery management system (BMS) function verification, combined with the design of automated equipment and standardized processes, realizes the efficient and accurate evaluation of retired batteries. Its test system architecture mainly includes a detection module, a control module, and a data analysis module; specifically as follows: Detection module: Covers the core detection links of the whole process, including odor and appearance inspection, safety performance testing (insulation performance, withstand voltage performance testing, and ultrasonic scanning internal structure analysis), state of health assessment (remaining capacity, internal resistance, voltage / temperature consistency analysis), and BMS function verification (SOC estimation error calibration, current / voltage measurement error detection), ensuring a comprehensive assessment from the outside to the inside, from performance to the system.

[0035] The detection module, as the basic data source of the system, is deeply linked with other modules through physical interfaces and signal transmission links: Odor and appearance inspection: After preliminary screening by manual or visual recognition equipment, qualified batteries enter the next link through a mechanical transmission device, and basic battery information (such as model, retirement source) is synchronized to the control module and the data analysis module by means of scanning codes, etc.

[0036] Safety performance test: Insulation and withstand voltage test: Detection equipment (such as insulation resistance tester, withstand voltage tester) is connected to the control module through the control line. The control module issues test instructions, and the test data is real-time fed back to the data analysis module.

[0037] Ultrasonic scanning internal structure analysis: A multi-band focused ultrasonic probe is used (5MHz to detect the electrolyte distribution, 20MHz to scan the electrode interface), and imaging is achieved through the difference in acoustic impedance; combined with a convolutional neural network (CNN) defect recognition model (the training data includes 100,000 industrial CT and ultrasonic comparison images, and the recognition accuracy is >95%), microscopic defect detection is realized. The detection accuracy of the lithium plating thickness reaches ±5μm, the positioning error of the electrolyte dry area is <1%, internal defect batteries can be eliminated in advance, and the thermal runaway accident rate after cascade utilization is reduced by 40%. Health status assessment (remaining capacity, internal resistance, etc.): Charging and discharging equipment, internal resistance testers, etc. are connected to the control module through communication interfaces (such as CAN bus, RS485). The control module automatically adjusts the charging and discharging rate and temperature (invoking a high and low temperature chamber), and synchronizes parameters such as voltage, current, and temperature to the data analysis module through a data acquisition device at a sampling interval of ≤1s for calculating core indicators such as remaining capacity and internal resistance growth rate.

[0038] BMS function verification: The battery BMS is docked with the control module through a special test fixture. The control module simulates battery operation signals (such as charging and discharging instructions) to verify parameters such as the SOC estimation error and voltage / current measurement accuracy of the BMS, and the results are directly transmitted to the data analysis module for functional evaluation.

[0039] Control module: Through automation control technology, precise control of charging and discharging equipment, data acquisition equipment, and environmental simulation devices (such as high and low temperature chambers) is achieved, supporting dynamic performance testing at different temperature environments and charging and discharging rates, and improving the standardization and controllability of the detection process.

[0040] The control module is connected to charging and discharging equipment, high and low temperature chambers, etc. through an industrial bus (such as PROFINET, Modbus) to achieve dynamic adjustment of detection parameters and equipment status monitoring, and undertake data preprocessing and process scheduling functions: Multi-scenario working condition simulation test bench: A programmable load simulator is developed to support random power fluctuations in the energy storage scenario (compliant with the GB / T 34131-2023 standard) and low-speed dynamic emergency acceleration / braking pulses. Record the voltage transient response at a sampling interval of 1ms and analyze the BMS communication delay; when the voltage recovery time >200ms in the pulse test, trigger the warning of the stripping of the electrode active material, and the screening error is reduced by 35%.

[0041] Process Automation: According to the instructions of the data analysis module, control the robotic arm or conveyor belt to guide the battery to specific workstations (such as secondary inspection or re-evaluation), realizing the "detection - scheduling" closed-loop.

[0042] Data Analysis Module: Integrate multi-dimensional data to build an evaluation model and generate a comprehensive report containing scenario suggestions: Data Aggregation and Processing: Receive detection data through the database interface, clean and associate it to establish a full life cycle file of the battery (such as QR code identification), and generate visual results such as charge and discharge curves and health trend charts.

[0043] Evaluation Logic and Model: Safety Performance Judgment: Batteries that do not meet the standards are directly marked as "disassembled and recycled".

[0044] Health Status Classification: Combine parameters such as remaining capacity and BMS accuracy to generate recommended reports such as "Grade A Energy Storage"; batteries with marginal indicators trigger dynamic environmental re-evaluation (such as capacity test at 50°C high temperature).

[0045] Deep Mining of Historical Data: Establish a "digital twin file" of the battery, analyze stress events such as deep discharge and high-temperature operation in vehicle network data, and calculate the stress accumulation index K = Σ (event intensity × frequency) / design life (deep discharge = 1.2, high-temperature operation = 0.8); when K>1.5, the life prediction model increases a 20% attenuation compensation coefficient, reducing the cycle life prediction error from ±10% to ±6%.

[0046] Scenario Adaptation Risk Scoring: Generate a score from 0 to 100 based on the formula R = 0.4×SOH + 0.3×(1 - |ΔV| / V_nom) + 0.2×(1 - K / 3) + 0.1×BMS_Accuracy: When R≥80 points, recommend energy storage power stations with a capacity attenuation rate of 8.2%±1.1% in 12 months; When 60≤R<80 points, recommend electric forklifts with an attenuation rate of 12.5%±2.3%; the dispersion of the attenuation rate of the comparison group reaches ±15%.

[0047] Process Optimization Feedback: Train a machine learning model through historical data (such as predicting the battery cycle life and optimizing the detection process timing), and feed back to the control module to adjust the detection strategy, continuously improving the system efficiency and accuracy.

[0048] Data Flow between Modules in the System: The detection module collects raw data → the control module preprocesses and transmits it → the data analysis module models and evaluates → the results are fed back to the control module and the detection module.

[0049] System function closed-loop: Through the "detection - control - analysis - feedback" cycle, the full-process automation from battery feeding to graded output is realized, avoiding efficiency losses and evaluation deviations caused by manual intervention, and finally achieving the goal of "efficient screening, precise grading, and safe utilization".

[0050] As Figure 1 shown, the method for evaluating the cascade utilization of power batteries using the above system includes the following steps: Step 1: Conduct appearance and odor inspections to screen batteries without damage, leakage, or abnormal odor. Step 2: Conduct safety performance tests in sequence. The safety performance tests include insulation, withstand voltage tests, and ultrasonic scanning for internal structure analysis to eliminate unqualified batteries. Step 3: Dynamic environment simulation. Through room temperature, high and low temperature charge and discharge, rate tests, and internal resistance analysis, the state of health is detected. At the same time, connect to a multi-scenario working condition simulation test bench to simulate the constant power charge and discharge of the energy storage scenario to fluctuate within ±10%, and make the pulse charge and discharge of the low-speed power scenario have a pulse width of 0.1 - 2s and a peak current of 3C. Record the voltage transient response time and fluctuation amplitude to evaluate the remaining capacity and consistency; the voltage consistency evaluation uses a high-precision multimeter with a resolution of 0.1mV and a measurement error of ≤±0.5mV; the temperature consistency is detected by a distributed temperature sensor with an accuracy of ±0.5°C and a sampling interval of ≤2s; in the high and low temperature charge and discharge tests, the high temperature environment is 50°C ± 1°C, the low temperature environment is -20°C ± 1°C, and 5 charge and discharge cycles are carried out at each temperature point. The average capacity attenuation rate is taken as the evaluation basis. Among them, when the low temperature discharge capacity attenuation rate ≤ 35% is judged as qualified. Step 4: Conduct BMS function verification to calibrate the SOC estimation and the measurement accuracy of voltage and current; synchronously execute historical data traceability: Analyze the charge and discharge log of the battery in the 100,000 kilometers before retirement, extract the deep discharge frequency with SOC < 10% and the proportion of high temperature operation time with temperature > 45°C, and calculate the battery stress accumulation index. Step 5: Integrate the detection data to generate a battery state of health report; combine the stress accumulation index with the working condition simulation data to generate a scenario adaptation risk score within the range of 0 - 100 points.

[0051] Among them, the ultrasonic scanning for internal structure analysis uses a multi-band focused ultrasonic probe with a central frequency of 5 - 20MHz. Batteries with micro-short circuits of the electrode sheet, lithium deposition thickness ≥ 50μm, or electrolyte dry area > 5% are unqualified batteries.

[0052] In the historical data traceability of Step 4: For batteries with a deep discharge frequency > 50 times and a high temperature operation time accounting for 20% of the total life, the stress accumulation index K ≥ 1.8; when K ≥ 2.0, generate a warning label of "Prohibited for high-rate scenarios".

[0053] In Step 5, the scene adaptation risk scoring calculation model is as follows: R = 0.4×SOH + 0.3×(1 - |ΔV| / V_nom) + 0.2×(1 - K / 3) + 0.1×BMS_Accuracy where SOH is the state of health ratio of remaining capacity, |ΔV| is the amplitude of the working condition simulated voltage fluctuation, K is the stress accumulation index, and BMS_Accuracy is the SOC estimation accuracy; When R ≥ 80 points, it is recommended for A-level energy storage scenarios; when 60 ≤ R < 80 points, it is recommended for low-speed power scenarios.

[0054] In the appearance and odor inspection, when using a visual recognition device for detection, the resolution of the industrial camera is ≥ 8 million pixels, the recognition accuracy of the defect detection algorithm for cracks and bulges reaches ±0.05 mm, and a gas sensor is used for leakage detection with a detection threshold of ≥ 500 ppm for the concentration of organic solvent gas.

[0055] In the insulation performance test, for single cells in energy storage scenarios, a DC voltage of 1500 V is applied for a test duration of 90 s, and the insulation resistance should be ≥ 100 MΩ; for battery modules in low-speed power scenarios, a DC voltage of 1000 V is applied for a test duration of 60 s, and the insulation resistance should be ≥ 50 MΩ.

[0056] In the voltage withstand test, the test voltage in the energy storage scenario is 1.8 times the rated voltage of the battery, and the test voltage in the low-speed power scenario is 1.5 times the rated voltage of the battery. The holding time is 60 s for both, and the leakage current thresholds are set to ≤ 0.5 mA for the energy storage scenario and ≤ 1 mA for the low-speed power scenario. The energy storage scenario and the low-speed power scenario are subjected to a voltage withstand test at 1.8 times the voltage and 1.5 times the voltage respectively (leakage current ≤ 0.5 mA / 1 mA) to achieve the verification of insulation strength grading, accurately identify insulation defects, increase the detection sensitivity by 30%, and improve the efficiency by 70% in combination with automated equipment.

[0057] The room temperature charge and discharge test is carried out in an environment of 25℃ ± 1℃, the charge and discharge rate is 0.5C, the remaining capacity is calculated by the ampere-hour integration method, and the error is ≤ 1.5%. In the voltage / temperature consistency evaluation, the voltage range of single cells is ≤ 50 mV, and the temperature range is ≤ 2℃. In the high and low temperature charge and discharge test, the high temperature environment is 50℃ ± 1℃, the low temperature environment is -20℃ ± 1℃, 5 charge and discharge cycles are carried out at each temperature point, and the average capacity decay rate is taken as the evaluation basis. Among them, when the low temperature discharge capacity decay rate is ≤ 35%, it is judged as qualified.

[0058] The internal resistance analysis uses the AC impedance method, with a test frequency range of 5 mHz - 50 kHz. The internal resistance is fitted through an equivalent circuit model, and the measurement accuracy is ≤ ±0.8% or ±0.05 mΩ (taking the larger value). The internal resistance growth rate is calculated based on the initial internal resistance, and the thresholds are ≤ 20% for the energy storage scenario and ≤ 30% for the low-speed power scenario.

[0059] In the BMS function verification, the SOC estimation error is calibrated by simulating charge and discharge conditions. The requirement for the energy storage scenario is ≤ 2%, and for the low-speed power scenario is ≤ 5%. The voltage measurement error is ≤ ±5 mV (for energy storage) or ±10 mV (for low-speed power), and the current measurement error is ≤ ±0.5% FS (for energy storage) or ±1% FS (for low-speed power).

[0060] During the data integration process, the control module collects data through the CAN bus with a communication cycle of ≤ 10 ms. The data integrity rate is ≥ 99.8%, and the SHA-256 algorithm is used to perform blockchain evidence preservation on the detection data to ensure the data cannot be tampered with.

[0061] In the health status detection, a high-precision multimeter is used for the voltage consistency assessment, with a resolution of 0.1 mV and a measurement error of ≤ ±0.5 mV. The temperature consistency is detected by a distributed temperature sensor, with an accuracy of ±0.5 °C and a sampling interval of ≤ 2 s.

[0062] In the dynamic environment simulation, the temperature control accuracy of the high and low temperature chamber is ±0.5 °C, the temperature change rate is ≥ 5 °C / min, and it can simulate extreme temperature conditions in the range of -30 °C to 60 °C, meeting the requirements of the GB / T34015.3 standard.

[0063] In the multi-scenario working condition simulation test bench: When simulating constant power fluctuations in the energy storage scenario, the voltage transient response time is ≤ 100 ms and the fluctuation amplitude is ≤ 2% of the rated voltage. When simulating pulse charge and discharge in the low-speed power scenario, the SOC estimation offset is ≤ 5%, and the BMS temperature sampling delay is ≤ 500 ms.

[0064] The data analysis module uses a machine learning model to predict the battery cycle life. The training data includes ≥ 100,000 groups of historical detection data, and the prediction error is ≤ 10%. The detection process priority is dynamically adjusted according to the prediction results. Using a machine learning model with ≥ 100,000 groups of data to predict the battery cycle life (error ≤ 10%), and dynamically adjusting the detection priority to achieve scientific life prediction and intelligent detection process, providing a quantitative basis for scenario matching, reducing ineffective detections, increasing the efficiency by more than 20%, and at the same time supporting full-life cycle performance tracking and maintenance optimization.

[0065] This method realizes the efficient and accurate evaluation of retired batteries through core technologies such as dynamic environment simulation, automated data collection, and multi-level safety threshold setting. Among them, dynamic environment simulation uses high and low temperature chambers to simulate extreme temperature environments in actual battery use (such as high temperature of 50°C and low temperature of -10°C), and systematically tests the performance degradation of the battery under different temperature gradients to ensure that the evaluation results are close to the actual application scenario; automated data collection records key parameters such as voltage, current, and temperature in real time at a high-frequency sampling interval of ≤1s, and constructs a complete battery performance database through a data continuous collection and dynamic calibration mechanism to provide a reliable basis for the health state evaluation; the multi-level safety threshold system sets different safety standards according to battery types (single cell / module / cluster) and application scenario differences. For example, the insulation resistance in the energy storage scenario is ≥100MΩ, and in the low-speed power scenario is ≥50MΩ, realizing refined control from safety performance to function adaptation.

[0066] The technical highlights are reflected in aspects such as the innovation of the multi-dimensional detection system, the integration of standardization and automation, the intelligent grading decision-making system, and the dynamic environment adaptability technology. This method first integrates safety performance (insulation, withstand voltage, ultrasonic scanning test), health state (remaining capacity / internal resistance / consistency analysis), and BMS function verification (SOC estimation error / voltage measurement accuracy) into a standardized evaluation process, covering the full life cycle requirements from the screening, grading to reuse of retired batteries, breaking through the limitations of traditional single-dimensional detection; based on national standards such as GB / T34015.3 to standardize detection conditions (such as charge and discharge rates, equipment accuracy), combined with automated charge and discharge equipment and environment simulation devices, to realize the standardization and high efficiency of the detection process, with an efficiency improvement of more than 50% compared to traditional manual operations; through the deep coupling of detection data and algorithm models, automatically generate battery grading reports (Grade A for energy storage, Grade B for low-speed power, Grade C for backup power, etc.), and dynamically match the echelon utilization scenarios according to safety thresholds and performance indicators, realizing the intelligent linkage of "detection - grading - application"; supporting complex test conditions such as high and low temperatures (-20°C to 50°C), large-rate charge and discharge (such as 1C, 3C), etc., by simulating the actual working conditions of scenarios such as energy storage and low-speed electric vehicles, improving the reliability and scenario adaptability of the evaluation results.

[0067] In terms of the advantageous effects, through multiple safety detections such as insulation resistance testing and withstand voltage test protection verification, this method accurately eliminates retired batteries with potential safety hazards, reducing the risk of secondary utilization by more than 60%, and significantly enhancing safety. The automated detection process reduces manual intervention by more than 70%, shortens the detection cycle from 48 hours of the traditional method to within 24 hours, and at the same time reduces test errors through standardized equipment, achieving a dual optimization of cost efficiency with a reduction in comprehensive cost by more than 30%. By establishing a hierarchical mechanism of "direct utilization for qualified ones - marginal reevaluation - directional downgraded use - disassembly and recycling", batteries that do not fully meet the standards (such as remaining capacity of 40% - 60%) are recommended for scenarios such as backup power supplies and small electronic devices, increasing the cascade utilization rate to 85%, reducing disassembly waste by 20% compared to the traditional screening mode, and maximizing resource utilization.

[0068] Integrating ultrasonic scanning for microscopic defect detection, dynamic working condition simulation, and BMS function verification, it breaks through the limitations of traditional single dimensions. The detection efficiency is increased by more than 50% compared to traditional manual operations. Based on standards and specifications such as GB / T34015.3, the data integrity rate ≥ 99.8% and is stored on the blockchain. Through differential safety thresholds (such as insulation resistance standards for energy storage and low-speed power) and risk scoring models, the application scenarios are accurately matched, and the cascade utilization rate is increased to 85%.

[0069] Embodiment 1 Battery evaluation for the energy storage scenario. In this embodiment, the battery evaluation is carried out according to the following steps: Appearance and odor inspection: Under good lighting conditions, visually inspect whether the battery appearance is intact, the surface is flat and dry, without damage, deformation, leakage, or abnormal odor, and eliminate the batteries that do not meet the requirements.

[0070] Safety performance test: Conduct insulation resistance testing (requirement ≥ 100 MΩ) and withstand voltage testing (no breakdown) on the batteries that pass the appearance screening in sequence, and directly eliminate the batteries that do not meet the standards.

[0071] Health status detection: Place the battery in an environment at room temperature (25°C ± 2°C), adjust the SOC to 50%, calculate the remaining capacity (required ≥ 70%) through charge and discharge testing, and analyze the internal resistance growth rate (≤ 20%) and voltage / temperature consistency.

[0072] BMS function verification: Simulate the working conditions of the energy storage scenario, verify the SOC estimation error (≤ 3%) and voltage / current measurement accuracy of the BMS, and ensure the accuracy of the system.

[0073] Generate a comprehensive report: Recommend the qualified batteries for large-scale energy storage systems; Mark the unqualified batteries as "pending reevaluation" and transfer them to other scenario processes.

[0074] Through this embodiment, safe batteries can be accurately screened: potential safety hazard batteries are eliminated through insulation / voltage-withstand testing to ensure the safe operation of the energy storage system; high-demand scenarios are adapted: the remaining capacity, internal resistance, and BMS accuracy are strictly limited to ensure the long cycle life and energy output stability of the batteries in the energy storage scenario; resource waste is reduced: unqualified batteries enter the re-evaluation process instead of being directly disassembled, increasing the possibility of secondary use.

[0075] Embodiment II For the battery evaluation in the low-speed power scenario, the battery evaluation is carried out according to the following steps in this embodiment: Appearance and odor inspection: Similar to Embodiment I, batteries with abnormal appearances are eliminated to ensure that the batteries entering the subsequent process have no visible physical damage.

[0076] Safety performance test: Conduct a short-circuit test (no fire within 10 minutes under simulated extreme conditions), and supplement the verification of overcharge / over-discharge protection functions to eliminate batteries with unqualified safety performance.

[0077] Health status detection: Place the battery in a low-temperature environment (-10°C ± 2°C), adjust the SOC to 80%, and test the low-temperature discharge capacity (≥60%) and voltage consistency (voltage difference ≤ 0.1V) to evaluate the performance stability of the battery in the low-temperature scenario.

[0078] BMS function verification: Focus on verifying the voltage measurement error (≤5%) and charge / discharge control logic of the BMS in the low-temperature environment to ensure reliable operation in low-speed power equipment.

[0079] Generate a comprehensive report: Qualified batteries are adapted to low-speed power scenarios such as electric bicycles and forklifts; unqualified batteries are transferred to the backup power scenario for re-evaluation.

[0080] This embodiment can strengthen the adaptation to extreme environments: Through low-temperature performance testing, low-speed power batteries suitable for cold regions are screened out, expanding the application scenarios; Ensure the reliability of power output: Focus on voltage consistency and the low-temperature adaptability of the BMS to reduce the energy fluctuation and control failure risks during equipment operation; Hierarchical re-evaluation mechanism: Unqualified batteries can be redirected to low-power scenarios to improve the overall resource utilization rate.

[0081] Embodiment III For the backup power scenario evaluation, the battery evaluation is carried out according to the following steps in this embodiment: Appearance and odor inspection: Priority is given to screening batteries with a complete appearance but marginal performance indicators (such as a remaining capacity of 40%-60%) to enter the backup power scenario evaluation process.

[0082] Safety performance re-inspection: Conduct an insulation resistance test (≥50MΩ) and a voltage-withstand test to ensure that the basic safety performance meets the requirements of the backup power scenario.

[0083] Health status detection: Test the remaining capacity (≥40%) and the internal resistance growth rate (≤30%) at room temperature, and relax the consistency requirements to adapt to low-power requirements.

[0084] BMS function verification: Simplify the verification process, and mainly confirm that the SOC estimation function is basically normal (error ≤10%), meeting the basic monitoring requirements of the backup power supply.

[0085] Generate a comprehensive report: The qualified batteries are recommended for scenarios such as backup power supplies for communication base stations and energy storage for street lights; the unqualified batteries enter the disassembly and recycling process.

[0086] This embodiment can revitalize batteries with marginal performance: Provide a feasible path for batteries with a relatively low remaining capacity but meeting safety standards, reducing the waste of resources that "can be used but not used"; Reduce application costs: Relax the consistency and BMS accuracy requirements, match the low-power and long standby requirements of the backup power supply scenario, and reduce the system integration cost; Control the safety bottom line: Ensure the basic safety of the backup power supply scenario through re-inspection, and avoid the low-power scenario becoming a safety blind spot.

[0087] Embodiment 4 Re-evaluation before disassembly and recycling. In this embodiment, the battery evaluation is carried out according to the following steps: Appearance and odor inspection: For batteries with a remaining capacity <40% or with serious damage to the appearance (such as a cracked housing), directly enter the re-evaluation process before recycling.

[0088] Safety performance confirmation: Quickly detect the insulation resistance (less than 10 MΩ is considered unqualified) and whether there is liquid leakage to confirm that there is no safety risk during disassembly.

[0089] Key parameter extraction: Through rapid internal resistance measurement and elemental analysis, identify the content and recycling value of recyclable materials (such as precious metals like lithium, cobalt, nickel, etc.).

[0090] Generate a comprehensive report: According to the material value evaluation results, it is recommended to enter the professional disassembly process to extract precious metal resources and handle harmful substances.

[0091] This embodiment can achieve double benefits in environmental protection and resources: By accurately judging the recycling value of batteries, prevent unusable batteries from flowing into the environment, and at the same time efficiently extract precious metals to alleviate the problem of resource shortage; Risk closed-loop management: Conduct safety confirmation on high-risk batteries (such as insulation failure) to prevent leakage or thermal runaway accidents during the disassembly process; Circular economy optimization: Establish a "detection-recycling" closed loop to enhance the scientific nature and sustainability of the full life cycle management of batteries.

[0092] Based on the evaluation methods of the above four embodiments, the evaluation conclusions and suggestions for power batteries based on the detection items are shown in Table 1.

[0093] Table 1

[0094] Example 5 Verification of internal structure by ultrasonic scanning. In this example, the analysis of the internal structure by ultrasonic scanning includes: for the batteries in the energy storage scenario, when the scanning resolution is ≥0.1 mm², it is judged as unqualified if the detected lithium deposition thickness is ≥80 μm or the density of micro-short circuit points is >3 per cm²; for the batteries in the low-speed power scenario, it is judged as unqualified if the detected dry electrolyte area is >8% or the wrinkled area of the electrode sheet is >5%.

[0095] Traditional battery safety detection can only identify macroscopic defects such as insulation failure and insufficient capacity, and cannot detect microscopic hidden dangers such as electrode lithium deposition and micro-short circuits, resulting in a relatively high risk of thermal runaway after cascade utilization. In this example, the accuracy and effectiveness of microscopic defect detection are verified through multi-band focused ultrasonic technology and CNN defect recognition model.

[0096] Test objects: 200 groups of retired ternary lithium batteries (from electric vehicles, with an average cycle life of 800 times and a remaining capacity of 65%-75% at retirement).

[0097] Test process: Detection by traditional method: Perform insulation resistance test (requirement ≥100 MΩ) and capacity test (qualified if the remaining capacity ≥70%).

[0098] Results: 15 groups were eliminated (5 groups with unqualified insulation and 10 groups with unqualified capacity), and the remaining 185 groups entered the ultrasonic scanning link.

[0099] Ultrasonic scanning detection: Equipment and parameters: Use a multi-band focused ultrasonic probe with a center frequency of 5-20 MHz. The 5 MHz probe detects the electrolyte distribution (resolution ≥0.1 mm²), and the 20 MHz probe scans the electrode interface (detects the lithium deposition thickness and micro-short circuit).

[0100] Algorithm support: A CNN model trained based on 100,000 industrial CT and ultrasonic comparison images automatically identifies lithium deposition (thickness ≥50 μm), micro-short circuit (density >3 per cm²) or dry electrolyte area (>5%).

[0101] Results: 32 groups of defective batteries were detected among the 185 groups, including 25 groups with lithium deposition (thickness 60-120 μm) and 7 groups with micro-short circuit points.

[0102] Disassembly verification: Manually disassemble the 32 groups of defective batteries, and confirm the internal state through optical microscopy and electron probe analysis.

[0103] Results: The detection results of 31 groups are consistent with the actual defects, and 1 group is misjudged (the electrolyte dry area is misjudged as lithium plating), with an accuracy rate of 98.4%.

[0104] Effect comparison: False judgment risk of traditional method: Among the 185 groups of batteries judged as "qualified" by the traditional method, 32 groups actually have microscopic defects. If directly used in cascade utilization (such as in energy storage scenarios), statistics after 6 months show that: The failure rate of the 32 groups of batteries without detected defects reaches 25% (thermal runaway or sudden capacity drop).

[0105] Advantages of ultrasonic scanning: By eliminating defective batteries in advance, this method realizes zero failures in the cascade utilization stage, and the thermal runaway accident rate is reduced by 40% compared with the traditional method.

[0106] Conclusion: Ultrasonic scanning technology can effectively identify internal microscopic defects missed by traditional methods, and significantly improve the detection accuracy in combination with the CNN algorithm, providing key safety guarantees for cascade utilization.

[0107] Example 6 Application of scenario adaptation risk score. Due to different historical usage conditions (such as deep discharge, high-temperature operation), the performance of retired batteries varies significantly in cascade utilization scenarios. The traditional random grouping method leads to a large dispersion in the capacity attenuation rate and low resource utilization. In this example, the scientificity of hierarchical application is verified through the stress accumulation index K and the scenario adaptation risk score R.

[0108] Test objects: 120 groups of retired lithium iron phosphate batteries (from online car-hailing, with a remaining capacity of 55%-75% at retirement), divided into three groups according to the scoring model: Group A (R≥80 points): 50 groups, recommended for energy storage power stations; Group B (60≤R<80 points): 70 groups, recommended for electric forklifts; Comparison group: 50 groups of batteries randomly grouped without scoring from the same batch.

[0109] Calculation of core parameters: Stress accumulation index K: Analyze vehicle networking data and extract key stress events within the first 100,000 kilometers before retirement: Frequency of deep discharge (SOC<10%): The average of Group A is 20 times, the average of Group B is 45 times, and the average of the comparison group is 38 times; Proportion of high-temperature operation duration (temperature>45℃): 8% in Group A, 18% in Group B, and 15% in the comparison group; Calculation formula: K = Σ(Event intensity × Frequency) / Design life (Design life is calculated based on 1000 cycles), where event intensity: Deep discharge = 1.2, High-temperature operation = 0.8.

[0110] Results: K of Group A = 0.96, K of Group B = 1.62, K of the control group = 1.35.

[0111] Scenario adaptation risk score R: Formula: R = 0.4×SOH + 0.3×(1 - |ΔV| / V_nom) + 0.2×(1 - K / 3) + 0.1×BMS_Accuracy Key parameters: SOH (State of Health ratio): 72% - 75% for Group A, 60% - 69% for Group B; |ΔV| / V_nom (Magnitude ratio of operating voltage fluctuation): ≤1.5% for Group A, ≤3.0% for Group B; BMS_Accuracy (SOC estimation accuracy): ≤2% for Group A, ≤5% for Group B.

[0112] Second-life utilization test: Group A (Energy storage power station): Application conditions: Constant temperature of 25°C, 0.5C charge and discharge, average annual cycle times of 300 times.

[0113] Data after 12 months: Capacity attenuation rate of 8.2% ± 1.1%, internal resistance growth rate of 18%, no thermal runaway events.

[0114] Group B (Electric forklift): Application conditions: -10°C to 40°C environment, pulsed charge and discharge (peak 3C, pulse width 0.5s), daily working 8 hours.

[0115] Data after 12 months: Capacity attenuation rate of 12.5% ± 2.3%, internal resistance growth rate of 28%, 2 groups with voltage consistency exceeding the standard (range > 100mV).

[0116] Control group (Random grouping): Randomly assigned to energy storage and forklift scenarios, not graded according to the score.

[0117] Data after 12 months: Capacity attenuation rate dispersion reaches ±15% (minimum 6%, maximum 21%), thermal runaway failure rate of 4%, and 30% of the batteries are prematurely taken out of use due to performance mismatch.

[0118] Effect analysis: Hierarchical Precision: By integrating multi-dimensional data such as SOH, voltage fluctuation, and stress history, the rating model reduces the fluctuation of the capacity attenuation rate in the same type of scenarios to within ±3% (Group A), significantly better than the ±15% of random grouping.

[0119] Resource Utilization: The attenuation rate of the batteries in Group B in the low-speed power scenario meets expectations, avoiding waste caused by direct elimination, and the overall echelon utilization rate is increased to 85% (only 65% by traditional methods).

[0120] Risk Control: The thermal runaway failure rate of the control group is 4%, while no similar accidents occurred in the rating and grading group, verifying the safety of the stress accumulation index matching the scenario.

[0121] Conclusion: The scenario-adapted risk rating model can achieve precise matching of echelon utilization scenarios based on battery historical data and real-time performance, reduce attenuation dispersion and safety risks, and provide a quantitative decision-making basis for the efficient reuse of retired batteries.

[0122] The present invention has made great breakthroughs in detection dimensions and technical means. Compared with the prior art, the health status and safety of batteries are mainly evaluated through appearance recognition, performance characteristic analysis (historical operation parameters + basic performance parameters), and internal microstructure detection (industrial CT, Li nuclear magnetic resonance imaging), with the focus on non-destructive detection of the internal structure, such as observing the swelling of electrode sheets through industrial CT and detecting the lithium fiber content by Li nuclear magnetic resonance. The present invention emphasizes a multi-dimensional detection system, integrating safety performance (insulation, withstand voltage), health status (remaining capacity, internal resistance, consistency), and BMS function verification (SOC estimation error, voltage / current measurement error), and introducing automated equipment and standardized processes (such as simulating the environment with a high and low temperature chamber and data acquisition with a sampling interval of ≤1 s). The present invention covers systematic detection of safety performance and BMS functions, while the prior art focuses more on microscopic analysis of the internal structure. The present invention emphasizes dynamic environment simulation (high and low temperatures, high-rate charge and discharge) and automated data acquisition, while the prior art ignores real-time dynamic testing and automated processes.

[0123] The present invention is significantly different from the prior art in evaluation logic and application scenarios. In the prior art, the safety of batteries is judged based on the results of internal structure detection (such as the state of electrode sheets, lithium fiber content), and the classification logic relies on qualitative analysis of health status and safety, without clearly associating with echelon utilization scenarios (such as energy storage, low-speed power, etc.). The present invention proposes intelligent grading decision-making, automatically generating a battery grading report (Grade A / B / C / D) based on detection data, and matching specific application scenarios (such as Grade A for energy storage, Grade B for low-speed electric vehicles), while providing re-evaluation or recycling suggestions for non-compliant batteries. The present invention constructs a scenario-customized grading system and realizes differential evaluation through multi-level safety thresholds (such as insulation resistance standards for different scenarios).

[0124] The present invention solves the problems of incomplete detection items, insufficient standardization, low efficiency, and weak data management in the prior art. By means of an automated process, the detection efficiency is increased by more than 50%, and the process is standardized based on national standards (such as GB / T34015.3).

[0125] The detection process of the present invention is more comprehensive and timely, taking into account both the process steps and the existing algorithm logic, and can complete more accurate battery evaluation while having a limited amount of calculation. In the prior art, the process includes preliminary sorting (industrial camera + bar code), performance detection (open circuit voltage + internal resistance), state of health assessment (mathematical model + internal resistance correction), safety assessment (thermal runaway + gas leakage), and intelligent classification (multi-objective optimization algorithm). The core lies in data quantitative analysis and intelligent classification, such as optimizing the fitness through a genetic algorithm. The present invention takes multi-dimensional physical detection as the core (safety performance, state of health, BMS function), combines dynamic environment simulation (high and low temperature, high rate), and automated equipment. The process focuses more on the comprehensive verification of physical performance rather than data model optimization. The present invention emphasizes the comprehensiveness of hardware detection (such as insulation and withstand voltage tests), compared with the prior art which focuses more on software algorithms and data management (such as thermal imaging, gas sensors, genetic algorithms). The present invention can reduce the dependence on algorithms through BMS function verification (such as SOC estimation error), and reduce the amount of calculation without affecting the overall evaluation accuracy.

[0126] The present invention specially sets up a safety assessment and classification logic. Compared with the safety assessment in the prior art which is based on the real-time monitoring of temperature gradient and gas leakage and triggers an isolation mechanism through a threshold, and the classification depends on a multi-objective optimization algorithm to comprehensively score performance, safety, and economic benefits. The safety assessment of the present invention is through standardized physical tests (such as insulation resistance ≥ 100 MΩ, no fire in short-circuit test), and the classification directly matches the scenario according to the detection result (such as remaining capacity ≥ 70% is Class A). The logic is simpler and relies on clear physical indicators. The safety assessment of the present invention relies more on predefined physical standards. Compared with the prior art which relies on real-time data and algorithms, the present invention can reduce the amount of calculation slightly without affecting the accuracy.

[0127] The present invention integrates three independent dimensions of safety performance, state of health, and BMS function into a unified evaluation process, rather than simply superimposing them, solving the problem of one-sided detection in the prior art. The battery evaluation in the prior art either only focuses on the internal structure or emphasizes data algorithms, and neither involves the combination of multi-dimensional physical detection and automated process.

[0128] The scenario-customized grading system of the present invention: based on the detection results, different safety thresholds are proposed to match the application scenarios (such as different standards for energy storage / low-speed power / backup power supply), rather than general screening. It needs to be designed specifically in combination with the characteristics of the second-life utilization scenario, which is not a conventional means in this field.

[0129] Dynamic environment simulation and automation of the present invention: By introducing a high and low temperature chamber, large rate charge and discharge testing, and data acquisition with a sampling interval of ≤1 s, it simulates real usage scenarios and improves data continuity. Such dynamic test conditions and automation control schemes have never been disclosed in the prior art.

[0130] Aiming at the problems of prominent safety hazards, low detection efficiency, and insufficient standardization in the prior art, the present invention systematically improves the reliability and efficiency of retired battery assessment through multi-dimensional detection, standardized processes, and automated equipment, filling the blank of the evaluation standard in the scenario of cascade utilization.

[0131] Through core innovations such as a multi-dimensional detection system, dynamic environment simulation, automated processes, and scenario customization and grading, the present invention solves the problems of one-sided detection, low efficiency, and prominent safety hazards in the prior art. Its technical solution is not a simple combination of comparative documents, but a systematic optimization for the cascade utilization scenario, with significant non-obviousness and creativity, providing a new technical path for the efficient and safe utilization of retired batteries.

[0132] The present invention can perform multi-dimensional defect correlation detection: Combining internal ultrasonic scanning (physical layer), operating condition dynamic response (performance layer), and historical stress analysis (data layer) to discover hidden failures that cannot be identified by a single dimension (such as high-frequency micro-shorts exposed in pulse tests); Cross-life cycle data fusion: For the first time, integrating the full-cycle data of "manufacturing - service - retirement" of the battery into the assessment, and quantifying the impact of historical damage on the remaining life through the stress accumulation index; Dynamic risk scoring model: Innovatively weighting and fusing traditional detection data (SOH, internal resistance), dynamic performance (voltage transient response), and historical damage (K index) to output a scenario-based risk score, solving the matching blind spot in the cascade utilization scenario.

[0133] The above are only embodiments of the present invention. Specific technical solutions and / or common knowledge such as characteristics well known in the art are not described in detail here. It should be noted that for those skilled in the art, without departing from the technical solution of the present invention, several deformations and improvements can still be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. A method for evaluating a second-life power battery, characterized in that It includes the following steps: Step 1: Conduct appearance and odor inspections to screen batteries without damage, leakage, or abnormal odors; Step 2: Conduct safety performance tests in sequence. The safety performance tests include insulation, withstand voltage tests, and ultrasonic scanning for internal structure analysis to eliminate unqualified batteries; Step 3: Dynamic environment simulation. Through room temperature and high and low temperature charge and discharge, rate tests, and internal resistance analysis, the state of health is detected. At the same time, connect to a multi-scenario working condition simulation test bench to simulate constant power charge and discharge in the energy storage scenario with a fluctuation within ±10%, and make the pulse charge and discharge in the low-speed power scenario have a pulse width of 0.1 - 2s and a peak current of 3C. Record the voltage transient response time and fluctuation amplitude to evaluate the remaining capacity and consistency; The voltage consistency evaluation uses a high-precision multimeter with a resolution of 0.1mV and a measurement error of ≤±0.5mV; The temperature consistency is detected by a distributed temperature sensor with an accuracy of ±0.5°C and a sampling interval of ≤2s; In the high and low temperature charge and discharge tests, the high temperature environment is 50°C ± 1°C, and the low temperature environment is -20°C ± 1°C. Five charge and discharge cycles are performed at each temperature point, and the average capacity decay rate is taken as the evaluation basis. Among them, when the low temperature discharge capacity decay rate ≤ 35% is judged as qualified; Step 4: Conduct BMS function verification to calibrate the SOC estimation and the measurement accuracy of voltage and current; Synchronously execute historical data traceability: Analyze the charge and discharge logs of the battery in the 100,000 kilometers before retirement, extract the deep discharge frequency with SOC < 10% and the proportion of high temperature operation duration with temperature > 45°C, and calculate the battery stress accumulation index; Step 5: Integrate the detection data to generate a battery state of health report; Combine the stress accumulation index with the working condition simulation data to generate a scenario adaptation risk score within the range of 0 - 100 points; In the dynamic environment simulation, the temperature control accuracy of the high and low temperature chamber is ±0.5°C, and the temperature change rate ≥ 5°C / min, which can simulate extreme temperature working conditions within the range of -30°C to 60°C.

2. The method according to claim 1, wherein In the ultrasonic scanning for internal structure analysis, a multi-band focused ultrasonic probe with a center frequency of 5 - 20MHz is used. Batteries with micro-shorts of the electrode sheet, lithium deposition thickness ≥ 50μm, or electrolyte dry area > 5% are judged as unqualified batteries.

3. The method for evaluating the cascade utilization of power batteries according to claim 1, characterized in that, In the BMS function verification, the SOC estimation error is calibrated by simulating charge and discharge working conditions. In the energy storage scenario, it should be ≤ 2%, and in the low-speed power scenario, it should be ≤ 5%; In the energy storage scenario, the voltage measurement error ≤ ±5mV, in the low-speed power scenario, the voltage measurement error ≤ ±10mV, in the energy storage scenario, the current measurement error ≤ ±0.5%FS, and in the low-speed power scenario, the current measurement error ≤ ±1%FS.

4. The stepwise utilization power battery evaluation method according to claim 1, wherein In Step 3, the room temperature charge and discharge test is carried out in an environment of 25°C ± 1°C, and the charge and discharge rate is 0.5C. The remaining capacity is calculated by the ampere-hour integration method with an error of ≤ 1.5%. In the evaluation of voltage and temperature consistency, the voltage range of single cells ≤ 50mV, and the temperature range ≤ 2°C.

5. The step utilization power battery evaluation method according to claim 1, characterized in that In Step 5, the calculation model of the scenario adaptation risk score is: R = 0.4×SOH + 0.3×(1 - |ΔV| / V_nom) + 0.2×(1 - K / 3) + 0.1×BMS_Accuracy Where SOH is the remaining capacity ratio, |ΔV| is the amplitude of the voltage fluctuation simulated under the working condition, K is the stress accumulation index, and BMS_Accuracy is the SOC estimation accuracy; When R ≥ 80 points, it is recommended for A-level energy storage scenarios; when 60 ≤ R < 80 points, it is recommended for low-speed power scenarios.

6. The ladder utilization power battery evaluation method according to claim 1, characterized in that In step two, in the insulation test, for single cells in the energy storage scenario, a DC voltage of 1500V is applied, the test duration is 90s, and the insulation resistance should be ≥ 100MΩ; for battery modules in the low-speed power scenario, a DC voltage of 1000V is applied, the test duration is 60s, and the insulation resistance should be ≥ 50MΩ.

7. The method for evaluating a cascade - utilized power battery according to claim 1, wherein, In step three, the internal resistance analysis uses the AC impedance method, the test frequency range is 5mHz - 50kHz, the internal resistance is fitted through an equivalent circuit model, the measurement accuracy is ≤ ±0.8% or ±0.05mΩ, taking the larger value, and the internal resistance growth rate is calculated based on the initial internal resistance, with the threshold being ≤ 20% for the energy storage scenario and ≤ 30% for the low-speed power scenario.

8. The evaluation method of the cascade utilization power battery according to claim 2, wherein The ultrasonic scanning internal structure analysis includes: for batteries in the energy storage scenario, the scanning resolution is ≥ 0.1mm², and it is determined to be unqualified when the detected lithium deposition thickness ≥ 80μm or the micro-short circuit point density > 3 per cm²; for batteries in the low-speed power scenario, it is determined to be unqualified when the detected electrolyte dry area > 8% or the electrode sheet wrinkling area > 5%.

9. The evaluation method of the cascaded utilization power battery according to claim 1, wherein, In the multi-scenario working condition simulation test bench: when simulating constant power fluctuation in the energy storage scenario, the voltage transient response time ≤ 100ms and the fluctuation amplitude ≤ 2% of the rated voltage; when simulating pulse charge and discharge in the low-speed power scenario, the SOC estimation offset ≤ 5%, and the BMS temperature sampling delay ≤ 500ms.

10. The secondary utilization power battery evaluation method according to claim 1, characterized in that, In the historical data traceability in step four: for batteries with a deep discharge frequency > 50 times and a high-temperature operation time reaching 20% of the total life, the stress accumulation index K ≥ 1.8; when K ≥ 2.0, a warning label of "Prohibited for use in high-rate scenarios" is generated.

Images